Abstract

A number of risk factors including the complement factor H (CFH) gene, smoking and Chlamydia pneumoniae have been associated with age-related macular degeneration (AMD). However, the mechanisms underlying how these risk factors might be involved in disease progression and disease aetiology is poorly understood. A cohort series of 233 individuals followed for AMD progression over a mean period of 7 years underwent a full eye examination, blood was taken for DNA and antibody titre and individuals completed a standard medical and general questionnaire. Y402H variants of the CFH gene were assessed with disease progression as well as examination of interaction between Y402H variants and smoking and Y402H variants and the pathogen C. pneumoniae. The CC risk genotype of Y402H was significantly associated with increased AMD progression [odds ratio (OR) 2.43, 95% confidence interval (95% CI) 1.07–5.49] as was smoking (OR 2.28, 95% CI 1.26–4.12). However, the risk of progression was greatly increased to almost 12-fold (OR 11.8, 95% CI 2.1–65.8) when, in addition to having the C risk allele, subjects also presented with the upper tertile of antibodies to the bacterial pathogen C. pneumoniae compared with those with the T allele of Y402H and the lowest antibody tertile. This demonstrates for the first time the existence of a gene environment–interaction between pathogenic load of C. pneumoniae and the CFH gene in the aetiology of AMD.

INTRODUCTION

Association of DNA variants from the complement factor H gene (CFH), in particular Y402H, with age-related macular degeneration (AMD) (ARMD1; MIM 603075) has now been convincingly demonstrated in many case–control studies (1–5). Moreover, a recent paper based on the Age-Related Eye Disease Study (AREDS) has also reported association of this variant with progression of AMD (6).

The CFH gene codes for complement factor H protein, a serum glycoprotein that acts as a key regulator of the alternative complement pathway. The alternative pathway constitutes one of the three pathways of the complement system representing the humoral component of the immune system’s natural defence against pathogenic infections. It is known to be activated by several factors including external pathogens. Thus, one hypothesis to explain AMD aetiology is that in genetically susceptible people, exposure to an organism could lead to an inappropriate, poorly regulated inflammatory response which ultimately results in AMD. Evidence to support this hypothesis comes from histopathological signs of chronic inflammation, disruption and the presence of macrophages, lymphocytes and mast cells at Bruch’s membrane (7). In addition, drusen also contain numerous proteins related to the process of inflammation or its aftermath (8–14), and some epidemiological studies have identified increased levels of the acute phase C-reactive protein, a systemic marker of inflammation in serum, in AMD patients (15,16).

Smoking presents as the most consistently modifiable environmental risk factor in AMD. However, other risk factors such as Chlamydia pneumoniae have also been implicated with two independent case–control studies demonstrating an association between exposure to C. pneumoniae infection and AMD using the enzyme-linked immuno-sorbent assay (ELISA) (17,18). We have also previously analysed C. pneumoniae antibody levels in a cohort of 254 AMD patients over a mean 7 year period and shown that an increased titre of antibody to C. pneumoniae was linked to more rapid progression of AMD between 2.1 (95% CI 0.92–4.69) and 3.0 (95% CI 1.46–6.37) times more risk of progression (19). In addition, C. pneumoniae has been reported to be found in five of nine surgically removed choroidal neovascular membrane (CNV) tissue from archival speciments of AMD eyes, but in none of 22 non-AMD eyes, using immunohistochemistry (IHC) and polymerase chain reaction (PCR) techniques. (20).

In the current study, we initially sought to explore the strength of the combined outcome of the Y402H CC risk genotype and the two environmental risk factors—smoking and exposure to chronic infection—as well as to access their interaction using our previously described AMD progression cohort (19).

RESULTS

A total of 254 unrelated individuals with early AMD at baseline examination were assessed for AMD progression in the course of this study as described previously (19). Genotyping information was obtained on 233 of these participants aged from 51 to 90 years, 77 of them showed signs of AMD progression, including 15 cases who progressed to end-stage disease. Of those 233 individuals with genotype information, three of these individuals had no serological data available. The group of 230 individuals was on average 3 years younger [71.3 (8.0 SD) versus 74.3 (7.2 SD), P 0.1], was followed for an average of 7.4 years as opposed to 7.2 years in the main group (P = 0.5) and showed no significant difference in smoking status (P = 1.0), study source (P = 0.4) or gender (P= 1.0). In those individuals included in the main study, the group of progressors had a mean age of 76.1 years (SD 7.5 years) compared with the mean age of the non-progressed group of 73.4 years (SD 7.0 years) (P = 0.008). There was no evidence of departure from Hardy–Weinberg equilibrium for the Y402H variant in progressors or non-progressors (P > 0.05).

The age and smoking defined as having ever smoked were strongly (P = 0.008 and P = 0.002, respectively) associated with AMD progression in this study (Table 1). The current study began before the high-dose nutritional supplement of antioxidants and zinc as recommended for treatment of advanced AMD results was published from the AREDS (21), and therefore these supplements were not used in this study population. However, the regular intake of vitamins was not significantly associated with the progression of early AMD in this study and was not therefore included in any subsequent analysis.

Table 1.

Analysis of the potential risk factors in the sample of 233 participants included in the analysis of the effect of the complement factor H on AMD progression

Characteristics  AMD status Univariate logistic regression analysis 
 Non-progressed Progressed OR 95% CI P-value 
 Number or mean Percentage or SD Number or mean Percentage or SD 
Total number of participants  156/233 67 77/233 33    
Age, years  73.4 7.04 76.1 7.47 1.06 1.01–1.1 0.008 
Time of follow-up, years  7.28 1.17 6.99 0.93 0.78 0.6–1.0 0.056 
Smoking: ever versus never No 95/125 76 30/125 24.0 2.4 1.4–4.3 0.002 
Yes 61/108 56.5 47/108 43.5 
Gender Male 73/108 67.6 35/108 32.4 1.1 0.6–1.8 0.847 
Female 83/125 66.4 42/125 33.6 
Source study VIP 92/131 70.2 39/131 29.8 1.4 0.8–2.4 0.229 
VECAT 64/102 62.7 38/102 37.3 
Characteristics  AMD status Univariate logistic regression analysis 
 Non-progressed Progressed OR 95% CI P-value 
 Number or mean Percentage or SD Number or mean Percentage or SD 
Total number of participants  156/233 67 77/233 33    
Age, years  73.4 7.04 76.1 7.47 1.06 1.01–1.1 0.008 
Time of follow-up, years  7.28 1.17 6.99 0.93 0.78 0.6–1.0 0.056 
Smoking: ever versus never No 95/125 76 30/125 24.0 2.4 1.4–4.3 0.002 
Yes 61/108 56.5 47/108 43.5 
Gender Male 73/108 67.6 35/108 32.4 1.1 0.6–1.8 0.847 
Female 83/125 66.4 42/125 33.6 
Source study VIP 92/131 70.2 39/131 29.8 1.4 0.8–2.4 0.229 
VECAT 64/102 62.7 38/102 37.3 

We analysed progressors and non-progressors for their Y402H genotype of the CFH gene. In the 233 genotype available individuals in the study, a higher proportion of individuals with the CC risk genotype of the CFH gene were identified in progressed cases 20/41 (48.8%) compared with 33/101 (32.7%) progressed cases with the CT genotype and 24/91 (26.4%) with the TT genotype (Table 2). Multivariate analysis adjusted for age, smoking, follow-up time, gender and study source indicated that those individuals with a CC genotype had a significantly increased risk of progression of disease [odds ratio (OR) 2.43, 95% confidence interval (95% CI) 1.07–5.49] compared with the TT reference genotype, whereas those with the TC genotype had an increased but non-significant trend for progression of disease [OR 1.40, 95% CI 0.72–2.71] (Table 2).

Table 2.

Logistic regression analysis of the association between CFH and progression of AMD

Genotype Number (%) of progressed cases Crude analysis Multivariate analysis, adjusted for age, smoking, follow-up time, gender and study source 
  OR 95% CI P-value OR 95% CI P-value 
TT 24/91(26.4) Reference   Reference   
TC 33/101(32.7) 1.35 0.73–2.53 0.34 1.4 0.72–2.71 0.32 
CC 20/41(48.8) 2.66 1.23–5.74 0.01 2.43 1.07–5.49 0.03 
Genotype Number (%) of progressed cases Crude analysis Multivariate analysis, adjusted for age, smoking, follow-up time, gender and study source 
  OR 95% CI P-value OR 95% CI P-value 
TT 24/91(26.4) Reference   Reference   
TC 33/101(32.7) 1.35 0.73–2.53 0.34 1.4 0.72–2.71 0.32 
CC 20/41(48.8) 2.66 1.23–5.74 0.01 2.43 1.07–5.49 0.03 

Several studies have now examined gene–environmental interactions of Y402H with smoking to investigate combined risks. In this study sample, 108 participants were former or current smokers. The rate of progression was 30/125 (24%) in the group who had never smoked compared with 47/108 (43.5%) in the group of those who have ever smoked. In multivariate analysis, adjusted for age, gender and follow-up time, those who ever smoked were at higher risk of AMD progression than those who had never smoked, with an OR of 2.28 (95% CI 1.26–4.12). In individuals with the CC genotype at Y402H, there was an increased risk of AMD progression associated with both non-smokers (7/20, 35%) and smokers (13/21, 61.9%). In the case of non-smokers, this risk was lower (OR 1.67, 95% CI 0.5–5.55) than that for smokers (OR 2.39, 95% CI 0.72–7.98). Although neither of these reached significance due to the low numbers in the stratified groups (Table 3), there was a clear indication that these two risk factors have a tendency to interact, producing the worst effect in smokers carrying the CC genotype. We tested for interaction between Y402H variants and smoking in relation to AMD progression. In our sample, relative risks (RRs) for AMD progression in the subgroups of subjects exposed to each (RR10 or RR01) or both (RR11) risk factors were RR11 = 2.83 (1.73–4.63), RR10 = 1.60 (0.80, 3.21) and RR01 = 1.78 (1.14, 2.79), where RR11 stands for risk ratio for progression in the group of smoking CC genotype carriers, RR10 is risk ratio for progression in the group of non-smoking CC genotype people and RR01 is risk ratio for progression in the group of smokers who do not have CC genotype. We estimated the relative excess risk (interaction contrast ratio or ICR) due to interaction to be 0.45, the attributable proportion (AP) due to interaction to be 0.16 and the synergy index (S) to be 3.91. All measures showed a positive departure from additivity compared with the sum of the separate effects of two risk factors, suggesting a positive interaction.

Table 3.

Logistic regression analysis of the association between CFH and progression of AMD stratified by smoking

Smoking status Genotype Number (%) Crude analysis progressed cases Multivariate analysis, adjusted for age, follow-up time, gender and study source 
   OR 95% CI P-value OR 95% CI P-value 
Never smokers TT 11/53 (20.7%) Reference   Reference   
TC 12/52 (23.1%) 1.14 0.45–2.89 0.77 1.18 0.45–3.12 0.73 
CC 7/20 (35.0%) 2.06 0.66–6.38 0.21 1.67 0.50–5.55 0.41 
Smokers (ever) TT 13/38 (34.2%) Reference   Reference   
TC 21/49 (42.9%) 1.44 0.6–3.47 0.41 1.46 0.57–3.73 0.42 
CC 13/21 (61.9%) 3.12 1.03–9.45 0.04 2.39 0.72–7.98 0.16 
Smoking status Genotype Number (%) Crude analysis progressed cases Multivariate analysis, adjusted for age, follow-up time, gender and study source 
   OR 95% CI P-value OR 95% CI P-value 
Never smokers TT 11/53 (20.7%) Reference   Reference   
TC 12/52 (23.1%) 1.14 0.45–2.89 0.77 1.18 0.45–3.12 0.73 
CC 7/20 (35.0%) 2.06 0.66–6.38 0.21 1.67 0.50–5.55 0.41 
Smokers (ever) TT 13/38 (34.2%) Reference   Reference   
TC 21/49 (42.9%) 1.44 0.6–3.47 0.41 1.46 0.57–3.73 0.42 
CC 13/21 (61.9%) 3.12 1.03–9.45 0.04 2.39 0.72–7.98 0.16 

We also investigated the combined effect of Y402H genotype and exposure to C. pneumoniae infection on AMD progression, which was found in our previous research to be independently associated with progression (19). The complete data on CFH genotypes and seroreactivity to C. pneumoniae were available in 230 participants. As bacterial titre increased from the lower to upper tertile of C. pneumoniae, there was a corresponding overall increase in the number of progressors from 16/77 (20.8%) in the lower tertile to 31/73 (42.5%) in the upper tertile of antibody titre (Table 4). Secondly, as the number of C risk alleles of the CFH gene increased, the number of progressors also increased from 24/90 (26.7%) in individuals with the TT genotype to 20/41(48.8%) in those with the CC genotype (Table 4). The most striking data are seen when the contrasting group with the TT genotype and lower level of antibody titre where 6/39 (15.4%) are compared with the group with the CC genotype and upper level of antibody titre where 8/12 (66.7%) showed AMD progression (Table 4).

Table 4.

AMD progression rates in the genotypes of CFH versus tertiles of antibodies to C. pneumoniae

CFH genotypes Tertiles of antibodies to C. pneumoniae, n (%) Total, n (%) 
 Tertile 1 Tertile 2 Tertile 3  
TT 6/39 (15.4) 8/30 (26.7) 10/21 (47.6) 24/90 (26.7) 
TC 4/25 (16.0) 15/34 (44.1) 13/40 (32.5) 32/99 (32.3) 
CC 6/13 (46.1) 6/16 (37.5) 8/12 (66.7) 20/41 (48.8) 
Total 16/77 (20.8) 29/80 (36.2) 31/73 (42.5) 76/230 (33.0) 
CFH genotypes Tertiles of antibodies to C. pneumoniae, n (%) Total, n (%) 
 Tertile 1 Tertile 2 Tertile 3  
TT 6/39 (15.4) 8/30 (26.7) 10/21 (47.6) 24/90 (26.7) 
TC 4/25 (16.0) 15/34 (44.1) 13/40 (32.5) 32/99 (32.3) 
CC 6/13 (46.1) 6/16 (37.5) 8/12 (66.7) 20/41 (48.8) 
Total 16/77 (20.8) 29/80 (36.2) 31/73 (42.5) 76/230 (33.0) 

In multivariate analysis adjusted for age, follow-up time, gender and study source, the most significant increase of AMD progression occurred in individuals with the highest tertile of antibody titre and the CC risk genotype at Y402H compared with individuals with the lowest tertile of antibody titre and the TT genotype at Y402H (OR 11.8, 95% CI 2.1–65.8) (Table 5). The risk of AMD progression was not as high, but still significantly increased in those with a combination of the upper tertile of antibody titre and the CC genotype at Y402H compared with the rest of the total sample (OR 4.5, 95% CI 1.2–16.6) (Table 5). When we compared progression in the group with the lowest C. pneumoniae antibody and the TT genotype (taking this group as a reference) to all other individuals of the sample, i.e. with middle or upper tertile of C. pneumoniae antibody and TC or CC genotype, there was also a significantly increased risk of progression in the latter group (OR 4.7, 95% CI 1.3–17.3) (Table 5).

Table 5.

Logistic regression analysis of the association between AMD progression and a combination of two AMD risk factors (CFH genotypes and exposure to C. pneumoniae)

  Number of progressed (%) ORa 95% CIa P-value ORb 95% CIb P-value 
Model 1 TT and lowest C. pneumoniae titre 6/39 (15.4) Reference   Reference   
CC and highest C. pneumoniae titre 8/12(66.7) 11.0 2.5–48.4 0.002 11.8 2.1–65.8 0.005 
Model 2 TT and lowest C. pneumoniae titre 6/39 (15.4) Reference      
All other combinations 70/191 (36.6) 3.0 1.2–7.5 0.02 2.9 1.1–7.5 0.03 
Model 3 All other combinations (reference) 68/218(31.2) Reference      
CC and highest C. pneumoniae titre 8/12(66.7) 4.5 1.3–15.2 0.02 4.7 1.3–17.3 0.02 
  Number of progressed (%) ORa 95% CIa P-value ORb 95% CIb P-value 
Model 1 TT and lowest C. pneumoniae titre 6/39 (15.4) Reference   Reference   
CC and highest C. pneumoniae titre 8/12(66.7) 11.0 2.5–48.4 0.002 11.8 2.1–65.8 0.005 
Model 2 TT and lowest C. pneumoniae titre 6/39 (15.4) Reference      
All other combinations 70/191 (36.6) 3.0 1.2–7.5 0.02 2.9 1.1–7.5 0.03 
Model 3 All other combinations (reference) 68/218(31.2) Reference      
CC and highest C. pneumoniae titre 8/12(66.7) 4.5 1.3–15.2 0.02 4.7 1.3–17.3 0.02 

aORs and CIs from the crude analyses.

bORs and 95% CIs adjusted for age, gender, smoking, follow-up time and study source.

We tested this sample for interaction between Y402H variants and C. pneumoniae antibody titre in relation to AMD progression. In our sample, RRs for AMD progression in the subgroups of subjects exposed to each (RR10 or RR01) or both (RR11) risk factors were RR11 (95% CI) = 2.59 (1.57–4.25), RR10 (95% CI) = 1.60 (0.95–2.71) and RR01 (95% CI) = 1.46 (0.95–2.26), where RR11 is risk ratio for progression in the group of those CC genotype carriers who have upper level of antibody titre, RR10 is risk ratio for progression in the group of CC genotype people who have low and medium levels of C. pneumoniae antibody titre and RR01 is risk ratio for progression in the group of participants with the upper level of antibody titre and no CC genotype. We estimated the ICR to be 0.53, the attributable proportion due to interaction to be 0.20 and the synergy index to be 5.76. All measures suggested a positive departure from additivity compared with the sum of the effects of each risk factor considered separately.

We also undertook additional analysis of the SNP rs2274700 in the CFH gene that has previously been demonstrated to show a greater significant association, compared with Y402H, with AMD in case–controls studies (22). We were also able to demonstrate that this SNP demonstrated a greater significant association with AMD progression (OR 4.6, 95% CI 1.7–12.4) than Y402H. Our analysis as to whether this SNP showed statistical evidence of a biological interaction with exposure to C. pneumoniae also demonstrated a positive departure from zero for a combined effect of the upper level of antibody titres and CC genotype of rs2274700 in the CFH gene, with the ICR, AP and S being 0.43, 0.17 and 6.58, respectively.

DISCUSSION

This study demonstrates for the first time an interaction between a pathogen, C. pneumoniae and the CFH gene and provides a possible hypothesis for the underlying mechanisms involved in AMD progression. It also provides further support for the CC risk variant of Y402H being associated with the progression of AMD as well as an associated risk through interaction with smoking. Numerous studies have now established that the CC risk variant of Y402H is associated with clinic-based case–control samples that are typically end- or late-stage in nature (1–5). However, it has only recently been demonstrated that association also exists for this gene with AMD disease progression. The findings presented here, showing an increased risk of disease with the CC genotype of Y402H, concur with those derived from the AREDS sample in the USA (6). A difference in baseline recruitment strategy and definitions between the two studies did not appear to influence the outcome, with both studies finding a similar association of the CC risk variant of Y402H and AMD progression (OR 2.43) in our study compared with an OR of 2.6 in the AREDS (6). In addition, we were also able to identify an increased risk in individuals with the CT genotype (OR 1.4), which was also similar to that reported in the AREDS (OR 1.6). This finding of an association of Y402H variants with AMD in two different progression studies from two different populations indicates that this finding is likely to reveal a true association.

In the current progression study, we observed (also previously reported) a two-fold risk of disease progression in individuals who had ever smoked compared with those who had never smoked, with OR 2.15, 95% CI 1.25–3.68 (19). The effect of smoking in our study was higher than that reported in the AREDS (OR 1.3, 95% CI 1.1–1.7) (6). Further analysis indicated that smoking maximally increased risk of progression if individuals were of the CC genotype with an OR of 2.39. To further explore whether there was any evidence of interaction between these two risk variables, we used three measures of interaction: ICR, AP and S (23,24). If there was no interaction between risk factors, then both the ICR and AP would be equal to zero and S would be equal to one. In the current study, all the three measures of ICR, AP and S indicated a positive departure from additivity of the risks imposed by the sum of the two risk factors if considered separately. Our findings on these three measures further strengthened the findings of an increased risk of AMD progression when both risk factors (CC genotype and current or past smoker) were considered together.

We have also previously shown an increase in AMD progression with increasing antibody titre to the bacterial pathogen of C. pneumoniae (19). Antibody titres to C. pneumoniae reflect chronic exposure to this organism. Seroreactivity to this organism is known to increase rapidly during childhood, with a very slow increase later in life (25,26). Although blood samples to measure seroreactivity to C. pneumoniae infection were not collected before the onset of AMD, as many of the participants were at a very mature age, seroreactivity to this organism would be relatively stable and thus would be unlikely to have a major effect on our findings. When we undertook stratification by CFH genotype and by C. pneumoniae antibody tertile, we found that individuals homozygous for the CC risk allele of this gene and in the upper tertile of antibody titre had the highest (11.8-fold) increased odds of disease progression compared with those individuals with a combination of the TT genotype of the CFH gene and the lowest C. pneumoniae antibody tertile. Given that the odds of progression of AMD based on CFH genotype identified in this study was 2.43 for the CC genotype and the odds of progression previously reported for the highest tertile of antibody titre to C. pneumoniae was 2.58 (definition 2) (19), this suggests that the interaction effect when considering both factors together may be super-additive. To further test this hypothesis, we again applied the three measures of interaction of ICR, AP and S. All the three measures of ICR, AP and S indicated a positive departure from additivity of the risks imposed by the sum of the two risk factors considered separately. Our findings on these three measures supported the findings of an increased risk of AMD progression when both risk factors (CC genotype and highest exposure to C.pneumoniae organism) were considered together.

In addition, replication of this superadditive effect with the SNP rs2274700 that has previously been described as a more significant SNP in the CFH gene for AMD in a case–control study (22) was also confirmed in this study with progression of AMD. Since both the Y402H variant and rs2274700 variant are in the same linkage disequilibrium (LD) block of the CFH gene, it suggests that the interaction effect is not restricted to a single SNP within this gene. Clearly, further work to extend this analysis to other SNPs in other LD blocks or in SNP haplotypes from this gene, as well as other genes implicated in AMD, is also warranted.

We are cautious not to over-estimate the meaning of these three measures of interaction in their prognostic value and their importance in interpreting the study results. This analysis of interaction is still a relatively new field and further work is required to define the significance of finding the excessive risk of the disease imposed by the combination of different risk factors in comparison with the additive sum of the risks from single exposures.

The findings presented in this paper go some way to establishing one potential mechanism of how CFH variants may play a role in the progression of AMD. Although we have tested only potential two-way interactions between CFH genotype and smoking and CFH genotype and C. pneumoniae antibody titre, it is clear that there may also be a three-way interaction between these risk terms but we are limited by our sample size and thus unable to supply sufficient statistical power to address this question. Replication of our findings is clearly warranted, but few studies have information on pathogen status including C. pneumoniae.

As a potential mechanism, we therefore propose that exposure to a pathogen similar to C. pneumoniae leads to activation of the alternative complement pathway. If this occurs in people with an altered CFH variant, there will be an inappropriately long and chronic activation of the complement system ultimately leading to end organ diseases, which manifests itself as AMD, and its progression to more serious stages.

Further work to explore the involvement of C. pneumoniae and other pathogens in AMD disease progression is necessary and if these findings can be verified, then it may offer the option of providing anti-microbial treatments to slow or prevent AMD progression.

MATERIALS AND METHODS

Subjects

The description of recruitment of participants into this study has been previously described. In brief, all participants were recruited from two Melbourne ophthalmic studies—who undertook a risk factor questionnaire, had anthropological and vascular measurements taken and were clinically examined as described previously (27). Stereo photography of the macula was performed using Kodachrome 64 ASA 35 mm slide films, following pupil dilatation with tropicamide (0.5%) and phenylephrine hydrochloride (10%). Photo grading was undertaken at baseline between 1992 and 1995 and again at follow-up, being an average of 7 years (range 6–9 years), and later in 2001–2002. All individuals were scored at baseline and at follow-up according to the international classification for AMD grading (28) using a six-level severity scale for AMD status: level 1, no drusen/hard drusen only; level 2, intermediate drusen or hyperpigmentation without hypopigmentation; level 3, soft distinct or indistinct drusen or pigmentary abnormalities (hyper- and hypopigmentation, or hypopigmentation alone); level 4, soft distinct or indistinct drusen and pigmentary abnormalities (hyper- or hypopigmentation); level 5, geographic atrophy; level 6, neovascular AMD. Participants at AMD levels 2–4 at baseline were included in the analysis of AMD progression. AMD progression included those cases in which there was an increase from level 2, 3 or 4 at baseline of one or more levels in either eye, together with cases in which there was an increase of two or more steps in the defined (as per the international classification for AMD grading) specific grades of size, total number, area occupied by a lesion and spread to a more central location within a level. The majority (82%) of the cases classified as progressed from baseline had AMD progression changing from one level to another level of grading in one or another eye. The remaining 18% (15 cases) progressed within a level (24). A more expansive and detailed description of progression in this study cohort is available (29).

The study protocol for this project was approved by the Human Research and Ethics Committee of the Royal Victorian Eye and Ear Hospital. Written informed consent was obtained from all individuals enrolled in the study. This study was conducted in accordance with the Declaration of Helsinki and following the National Health and Medical Research Council of Australia’s statement on ethical conduct in research involving humans, revised in 1999.

DNA analysis

Genomic DNA was isolated from venous blood leukocytes, using a standard phenol/chloroform extraction procedure. Genotyping was performed using the MassARRAY® platform (SEQUENOM) through the Australian Genome Research Facility, Brisbane, Australia, as described previously (30). In brief, 2.5 ng of genomic DNA was amplified using the PCR. For the T→C substitution at nucleotide position 1277 of exon 9 (rs1061170) (Y402H) of the CFH gene, the primers 5′acgttggatggttatggtccttaggaaaatg3′ (forward) and 5′acgttggatggcaacgtctatagatttaccc3′ (reverse) were used to give a PCR product of 97 bp. In the case of the SNP rs2274700 (Ala473Ala), the primers 5′ acgttggatggtttcaccatctgctgttac3′ (forward) and 5′acgttggatggaatctcagtatacatatgcc3″ (reverse) were used to give a PCR product of 100 bp. A MassEXTEND® reaction was initiated by adding DNA polymerase, dNTPs, ddNTPs and an extension primer (5′ctgtacaaactttcttccat3′ for Y402H and 5′catatgccttaaaagaaaaagc3′ for rs2274700) that allowed a 1 bp primer extension of either allelic variant at the polymorphic site. Two allele-specific products of different mass were generated and run through Maldi-TOF and genotypes simultaneously called using SpectroTYPER™ RT software. Validation of results was undertaken by performing unidirectional di-deoxy sequencing on a representative set of clinical samples (30).

C. pneumoniae serology

Plasma samples were collected in plastic cryo-tubes at −80°C, stored and analysed at the Department of Molecular Sciences, University of Tennessee Health Sciences Center (Memphis, TN, USA). ELISA tests were performed using elementary bodies (EBs) from C. pneumoniae AR39, and titres were expressed as optical density (OD) units. The C. pneumoniae whole organisms (AR39) were grown in HeLa cells, and the infectious stage of the organism (EBs) was harvested and stored at −80°C. The complete details of the ELISA test were reported previously (17,31,32). Briefly, the immunolon 2 plates (Dynex Technologies) coated with 0.5 µg antigen in phosphate-buffered saline (PBS) for 48 h at 4°C were incubated with a 1:250 dilution of patient sera in PBS. Then the plates were incubated with the alkaline phosphatase-conjugated goat anti-human IgG (Jackson Immunoresearch Laboratories, West Grove, PA, USA) and the substrate, p-nitrophenylphosphate (SigmaFAST tablets; Sigma Chemical Co.). Absorbance was read as OD units at 405 nm on a Perkin Elmer HTS 7000 Bio Assay Reader. For each serum sample, the OD value of a phosphate-buffered saline-coated well that had no antigen (antigen blank) was subtracted from the values for all the test wells for that antigen. Triplicate-blanked test OD values were averaged and reported for each patient. Laboratory personnel were masked to clinical and demographic information on the patients. Our analysis of the association between AMD progression and exposure to C. pneumoniae in this sample has been published (19).

Statistical analysis

Data were analysed using the χ2 test for categorical variables in univariate analysis to determine association of potential risk factors with progression of AMD. Allelic association of the Y402H and the rs2274700 variants of the CFH gene with AMD progression were quantified in multivariate logistic regression analysis, controlling for age and smoking, both risk factors for AMD and its progression. Information on smoking status was collapsed into two categories for inclusion in a multivariate model: ‘never smoked’ or ‘having ever smoked’. Most of those ‘having ever smoked’ were long-term smokers (29). The averages of the logged antibody titres for C. pneumoniae in the three groups were compared to detect a trend for an association. Multivariate logistic regression analysis was used to determine the strength of the association between the antibody titres to C. pneumoniae and AMD progression. Additional analysis to identify tendency to biological interaction was also undertaken using risk ratios and three measures of interaction proposed by Rothman and Greenland (23)—the ICR, AP and S. In the analyses, RRs for AMD progression in the presence of risk factors or their combination were calculated with those jointly unexposed as a reference category. An interaction effect is proved if the relative excess risk among those with combined exposure exceeds the sum of the relative excess risks for each of the component causes. Since ORs can overestimate the interaction effect when the prevalence of outcome is high and produce misleading results (24), we used only risk ratios, calculated with the software written and provided by Rothman (33). The measures of additive interaction were defined as follows (23,24):  

1
formula
 
2
formula
 
3
formula
where RR11, or R11/R00, stands for risk ratio comparing the outcome in the group of doubly exposed to both risk factors in comparison with the reference group of doubly unexposed subjects; RR10 stands for risk ratio comparing the outcome in subjects with the first risk factor in the absence of the second risk factor, with the reference group; RR01 stands for risk ratio comparing the outcome in subjects exposed to the second risk factor and unexposed to the first, with the reference group.

All analyses were performed in SPSS version 12.0.1 (SPSS Inc., Chicago, IL, USA). A 5% level of significance (two sided) was used.

FUNDING

This project was supported by the National Health and Medical Research Council of Australia through a Clinical Fellowship to R.H.G. and NHMRC grant 128201, the JACOM Foundation, the Ophthalmic Research Institute of Australia, Perpetual Trustees Australia Ltd (Ramaciotti Foundation), ANZ Executors & Trustee Company Ltd (The Hugh D.T. Williamson Foundation under the Medical Research and Technology in Victoria Program), The Royal Victorian Institute for the Blind, The Royal Victorian Eye and Ear Hospital Research Committee, The Lions Club of Victoria and The Australian Institute of Health and Welfare.

ACKNOWLEDGEMENTS

Dr Sadia Mahdi and Professor Gerald Byrne from the Tennessee University conducted the serological tests. We thank Professor Kenneth Rothman for helpful discussions.

Conflict of Interest statement. All the authors had full access to all of the data in the study and jointly take responsibility for the integrity of the data and the accuracy of the data analysis. None of the authors have any conflicts of interest or financial interests in the findings from this manuscript.

REFERENCES

1
Hageman
G.S.
Anderson
D.H.
Johnson
L.V.
Hancox
L.S.
Taiber
A.J.
Hardisty
L.I.
Hageman
J.L.
Stockman
H.A.
Borchardt
J.D.
Gehrs
K.M.
, et al.  . 
A common haplotype in the complement regulatory gene factor H (HF1/CFH) predisposes individuals to age-related macular degeneration
Proc. Natl Acad. Sci. USA
 , 
2005
, vol. 
102
 (pg. 
7227
-
7232
)
2
Scholl
H.P.
Fleckenstein
M.
Charbel Issa
P.
Keilhauer
C.
Holz
F.G.
Weber
B.H.
An update on the genetics of age-related macular degeneration
Mol. Vis.
 , 
2007
, vol. 
13
 (pg. 
196
-
205
)
3
Klein
R.J.
Zeiss
C.
Chew
E.Y.
Tsai
J.Y.
Sackler
R.S.
Haynes
C.
Henning
A.K.
SanGiovanni
J.P.
Mane
S.M.
Mayne
S.T.
, et al.  . 
Complement factor H polymorphism in age-related macular degeneration
Science
 , 
2005
, vol. 
308
 (pg. 
385
-
389
)
4
Haines
J.L.
Hauser
M.A.
Schmidt
S.
Scott
W.K.
Olson
L.M.
Gallins
P.
Spencer
K.L.
Kwan
S.Y.
Noureddine
M.
Gilbert
J.R.
, et al.  . 
Complement factor H variant increases the risk of age-related macular degeneration
Science
 , 
2005
, vol. 
308
 (pg. 
419
-
421
)
5
Edwards
A.O.
Ritter
R.
III
Abel
K.J.
Manning
A.
Panhuysen
C.
Farrer
L.A.
Complement factor H polymorphism and age-related macular degeneration
Science
 , 
2005
, vol. 
308
 (pg. 
421
-
424
)
6
Seddon
J.M.
Francis
P.J.
George
S.
Schultz
D.W.
Rosner
B.
Klein
M.L.
Association of CFH Y402H and LOC387715 A69S with progression of age-related macular degeneration
JAMA
 , 
2007
, vol. 
297
 (pg. 
1793
-
1800
)
7
Penfold
P.L.
Killingsworth
M.C.
Sarks
S.H.
Senile macular degeneration: the involvement of immunocompetent cells
Graefes Arch. Clin. Exp. Ophthalmol.
 , 
1985
, vol. 
223
 (pg. 
69
-
76
)
8
Hageman
G.S.
Mullins
R.F.
Russell
S.R.
Johnson
L.V.
Anderson
D.H.
Vitronectin is a constituent of ocular drusen and the vitronectin gene is expressed in human retinal pigmented epithelial cells
FASEB J.
 , 
1999
, vol. 
13
 (pg. 
477
-
484
)
9
Mullins
R.F.
Russell
S.R.
Anderson
D.H.
Hageman
G.S.
Drusen associated with aging and age-related macular degeneration contain proteins common to extracellular deposits associated with atherosclerosis, elastosis, amyloidosis, and dense deposit disease
FASEB J.
 , 
2000
, vol. 
14
 (pg. 
835
-
846
)
10
Johnson
L.V.
Ozaki
S.
Staples
M.K.
Erickson
P.A.
Anderson
D.H.
A potential role for immune complex pathogenesis in drusen formation
Exp. Eye Res.
 , 
2000
, vol. 
70
 (pg. 
441
-
449
)
11
Johnson
L.V.
Leitner
W.P.
Staples
M.K.
Anderson
D.H.
Complement activation and inflammatory processes in drusen formation and age related macular degeneration
Exp. Eye Res.
 , 
2001
, vol. 
73
 (pg. 
887
-
896
)
12
Anderson
D.
Ozaki
S.
Nealon
M.
Neitz
J.
Mullins
R.
Hageman
G.
Johnson
L.V.
Local cellular sources of apolipoprotein E in the human retina and retinal pigmented epithelium: implications for the process of drusen formation
Am. J. Ophthalmol.
 , 
2001
, vol. 
131
 (pg. 
767
-
781
)
13
Hageman
G.S.
Luthert
P.J.
Victor Chong
N.H.
Johnson
L.V.
Anderson
D.H.
Mullins
R.F.
An integrated hypothesis that considers drusen as biomarkers of immune-mediated processes at the RPE-Bruch’s membrane interface in aging and age-related macular degeneration
Prog. Retin. Eye Res.
 , 
2001
, vol. 
20
 (pg. 
705
-
732
)
14
Crabb
J.W.
Miyagi
M.
Gu
X.
Shadrach
K.
West
K.A.
Sakaguchi
H.
Kamei
M.
Hasan
A.
Yan
L.
Rayborn
M.E.
, et al.  . 
Drusen proteome analysis: an approach to the etiology of age-related macular degeneration. (Comment)
Proc. Natl Acad. Sci. USA
 , 
2002
, vol. 
99
 (pg. 
14682
-
14687
)
15
Seddon
J.M.
Gensler
G.
Milton
R.C.
Klein
M.L.
Rifai
N.
Association between C-reactive protein and age-related macular degeneration
JAMA
 , 
2004
, vol. 
291
 (pg. 
704
-
710
)
16
Vine
A.
Stader
J.
Branham
K.
Musch
D.
Swaroop
A.
Biomarkers of cardiovascular disease as risk factors for age-related macular degeneration
Ophthalmology
 , 
2005
, vol. 
112
 (pg. 
2076
-
2080
)
17
Kalayoglu
M.
Galvan
C.
Mahdi
O.
Byrne
G.
Mansour
S.
Serological association between Chlamydia pneumoniae infection and age-related macular degeneration
Arch. Ophthalmol.
 , 
2003
, vol. 
121
 (pg. 
478
-
482
)
18
Ishida
O.
Oku
H.
Ikeda
T.
Nishimura
M.
Kawagoe
K.
Nakamura
K.
Is Chlamydia pneumoniae infection a risk factor for age related macular degeneration?
Br. J. Ophthalmol.
 , 
2003
, vol. 
87
 (pg. 
523
-
524
)
19
Robman
L.
Mahdi
O.
McCarty
C.
Dimitrov
P.
Tikellis
G.
McNeil
J.
Byrne
G.
Taylor
H.
Guymer
R.
Exposure to Chlamydia pneumoniae infection and progression of age-related macular degeneration
Am. J. Epidemiol.
 , 
2005
, vol. 
161
 (pg. 
1013
-
1019
)
20
Kalayoglu
M.
Bula
D.
Arroyo
J.
Gragoudas
E.
D’Amico
D.
Miller
J.
Identification of Chlamydia pneumoniae within human choroidal neovascular membranes secondary to age-related macular degeneration
Graefes. Arch. Clin. Exp. Ophthalmol.
 , 
2005
, vol. 
243
 (pg. 
1080
-
1090
)
21
Bressler
N.M.
Bressler
S.B.
Congdon
N.G.
Ferris
F.L.
III
Friedman
D.S.
Klein
R.
Lindblad
A.S.
Milton
R.C.
Seddon
J.M.
Potential public health impact of Age-Related Eye Disease Study results: AREDS report no. 11
Arch. Ophthalmol.
 , 
2003
, vol. 
121
 (pg. 
1621
-
1624
)
22
Li
M.
Atmaca-Sonmez
P.
Othman
M.
Branham
K.E.
Khanna
R.
Wade
M.S.
Li
Y.
Liang
L.
Zareparsi
S.
Swaroop
A.
, et al.  . 
CFH haplotypes without the Y402H coding variant show strong association with susceptibility to age-related macular degeneration
Nat. Genet.
 , 
2006
, vol. 
38
 (pg. 
1049
-
1054
)
23
Rothman
K.
Greenland
S.
Modern Epidemiology
 , 
1998
Philadelphia, PA
Lipcott Williams and Wilkins
24
Kalilani
L.
Atashili
J.
Measuring additive interaction using odds ratios
Epidemiol. Perspect. Innov.
 , 
2006
, vol. 
3
 pg. 
5
 
25
Grayston
J.
Chernesky
M.
Clinical trials of antibiotic treatment of atherosclerotic disease
2006
Proceedings of the Eleventh International Symposium on Human Chlamydial Infection
Niagara-on-the-Lake, Ontario, Canada
(pg. 
621
-
630
)
26
Siscovick
D.S.
Schwartz
S.M.
Caps
M.
Wang
S.P.
Grayston
J.T.
Chlamydia pneumoniae and atherosclerotic risk in populations: the role of seroepidemiology
J. Infect. Dis.
 , 
2000
, vol. 
181
 
Suppl. 3
(pg. 
S417
-
S420
)
27
Robman
L.
McNeil
J.
Dimitrov
P.
Dowrick
A.
Tikellis
G.
Nicolas
C.
Cameron
J.
Guymer
R.
McGrath
B.
McCarty
C.
Methodology of the Cardiovascular Health and Age-Related Maculopathy (CHARM) Study
Ophthalmic Epidemiol.
 , 
2004
, vol. 
11
 (pg. 
161
-
179
)
28
Bird
A.C.
Bressler
N.M.
Bressler
S.B.
Chisholm
I.H.
Coscas
G.
Davis
M.D.
de Jong
P.T.
Klaver
C.C.
Klein
B.E.
Klein
R.
, et al.  . 
An international classification and grading system for age-related maculopathy and age-related macular degeneration. The International ARM Epidemiological Study Group
Surv. Ophthalmol.
 , 
1995
, vol. 
39
 (pg. 
367
-
374
)
29
Tikellis
G.
Robman
L.D.
Dimitrov
P.
Nicolas
C.
McCarty
C.A.
Guymer
R.H.
Characteristics of progression of early age-related macular degeneration: the Cardiovascular Health and Age-related maculopathy Study
Eye
 , 
2007
, vol. 
21
 (pg. 
169
-
176
)
30
Baird
P.N.
Islam
F.M.
Richardson
A.J.
Cain
M.
Hunt
N.
Guymer
R.
Analysis of the Y402H variant of the complement factor H gene in age-related macular degeneration
Invest. Ophthalmol. Vis. Sci.
 , 
2006
, vol. 
47
 (pg. 
4194
-
4198
)
31
Mahdi
O.
Horne
B.
Peeling
R.
Mullen
K.
Dillon
E.
Muhlestein
J.
Byrne
G.
Serological response to Clamydia pneumoniae and heat shock proteins in sera from patients with atherosclerosis
2002
Proceedings of the Tenth International Symposium on Human Chlamydial Infections
Antlaya, Turkey
(pg. 
329
-
332
)
32
Mahdi
O.S.
Horne
B.D.
Mullen
K.
Muhlestein
J.B.
Byrne
G.I.
Serum immunoglobulin G antibodies to chlamydial heat shock protein 60 but not to human and bacterial homologs are associated with coronary artery disease
Circulation
 , 
2002
, vol. 
106
 (pg. 
1659
-
1663
)
33
Rothman
K.
‘Episheet’ epidemiologic software
2005
 
version of November 10